An adaptive spatiotemporal correlation filtering visual tracking method.
Discriminative correlation filter (DCF) tracking algorithms are commonly used for visual tracking. However, we observed that different spatio-temporal targets exhibit varied visual appearances, and most DCF-based trackers neglect to exploit this spatio-temporal information during the tracking proces...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Public Library of Science (PLoS)
2023-01-01
|
| Series: | PLoS ONE |
| Online Access: | https://doi.org/10.1371/journal.pone.0279240 |
| _version_ | 1827986439460093952 |
|---|---|
| author | Yuhan Liu He Yan Wei Zhang Mengxue Li Lingkun Liu |
| author_facet | Yuhan Liu He Yan Wei Zhang Mengxue Li Lingkun Liu |
| author_sort | Yuhan Liu |
| collection | DOAJ |
| description | Discriminative correlation filter (DCF) tracking algorithms are commonly used for visual tracking. However, we observed that different spatio-temporal targets exhibit varied visual appearances, and most DCF-based trackers neglect to exploit this spatio-temporal information during the tracking process. To address the above-mentioned issues, we propose a three-way adaptive spatio-temporal correlation filtering tracker, named ASCF, that makes fuller use of the spatio-temporal information during tracking. To be specific, we extract rich local and global visual features based on the Conformer network, establish three correlation filters at different spatio-temporal locations during the tracking process, and the three correlation filters independently track the target. Then, to adaptively select the correlation filter to achieve target tracking, we employ the average peak-to-correlation energy (APCE) and the peak-to-sidelobe ratio (PSR) to measure the reliability of the tracking results. In addition, we propose an adaptive model update strategy that adjusts the update frequency of the three correlation filters in different ways to avoid model drift due to the introduction of similar objects or background noise. Extensive experimental results on five benchmarks demonstrate that our algorithm achieves excellent performance compared to state-of-the-art trackers. |
| first_indexed | 2024-04-09T23:31:24Z |
| format | Article |
| id | doaj.art-ff9036f832ec46e2b581c3c64fb4e7af |
| institution | Directory Open Access Journal |
| issn | 1932-6203 |
| language | English |
| last_indexed | 2024-04-09T23:31:24Z |
| publishDate | 2023-01-01 |
| publisher | Public Library of Science (PLoS) |
| record_format | Article |
| series | PLoS ONE |
| spelling | doaj.art-ff9036f832ec46e2b581c3c64fb4e7af2023-03-21T05:31:50ZengPublic Library of Science (PLoS)PLoS ONE1932-62032023-01-01181e027924010.1371/journal.pone.0279240An adaptive spatiotemporal correlation filtering visual tracking method.Yuhan LiuHe YanWei ZhangMengxue LiLingkun LiuDiscriminative correlation filter (DCF) tracking algorithms are commonly used for visual tracking. However, we observed that different spatio-temporal targets exhibit varied visual appearances, and most DCF-based trackers neglect to exploit this spatio-temporal information during the tracking process. To address the above-mentioned issues, we propose a three-way adaptive spatio-temporal correlation filtering tracker, named ASCF, that makes fuller use of the spatio-temporal information during tracking. To be specific, we extract rich local and global visual features based on the Conformer network, establish three correlation filters at different spatio-temporal locations during the tracking process, and the three correlation filters independently track the target. Then, to adaptively select the correlation filter to achieve target tracking, we employ the average peak-to-correlation energy (APCE) and the peak-to-sidelobe ratio (PSR) to measure the reliability of the tracking results. In addition, we propose an adaptive model update strategy that adjusts the update frequency of the three correlation filters in different ways to avoid model drift due to the introduction of similar objects or background noise. Extensive experimental results on five benchmarks demonstrate that our algorithm achieves excellent performance compared to state-of-the-art trackers.https://doi.org/10.1371/journal.pone.0279240 |
| spellingShingle | Yuhan Liu He Yan Wei Zhang Mengxue Li Lingkun Liu An adaptive spatiotemporal correlation filtering visual tracking method. PLoS ONE |
| title | An adaptive spatiotemporal correlation filtering visual tracking method. |
| title_full | An adaptive spatiotemporal correlation filtering visual tracking method. |
| title_fullStr | An adaptive spatiotemporal correlation filtering visual tracking method. |
| title_full_unstemmed | An adaptive spatiotemporal correlation filtering visual tracking method. |
| title_short | An adaptive spatiotemporal correlation filtering visual tracking method. |
| title_sort | adaptive spatiotemporal correlation filtering visual tracking method |
| url | https://doi.org/10.1371/journal.pone.0279240 |
| work_keys_str_mv | AT yuhanliu anadaptivespatiotemporalcorrelationfilteringvisualtrackingmethod AT heyan anadaptivespatiotemporalcorrelationfilteringvisualtrackingmethod AT weizhang anadaptivespatiotemporalcorrelationfilteringvisualtrackingmethod AT mengxueli anadaptivespatiotemporalcorrelationfilteringvisualtrackingmethod AT lingkunliu anadaptivespatiotemporalcorrelationfilteringvisualtrackingmethod AT yuhanliu adaptivespatiotemporalcorrelationfilteringvisualtrackingmethod AT heyan adaptivespatiotemporalcorrelationfilteringvisualtrackingmethod AT weizhang adaptivespatiotemporalcorrelationfilteringvisualtrackingmethod AT mengxueli adaptivespatiotemporalcorrelationfilteringvisualtrackingmethod AT lingkunliu adaptivespatiotemporalcorrelationfilteringvisualtrackingmethod |